The Chande Momentum Oscillator Explained: RSI Without the Smoothing

Pure momentum on a ±100 scale, and how it differs from RSI

Chande Momentum Oscillator
Chande Momentum Oscillator

The Chande Momentum Oscillator (CMO) is what you get when you take the core idea behind RSI — compare up-moves to down-moves — and strip out the smoothing that makes RSI sluggish. The result is a faster, more responsive momentum gauge that swings across a symmetric ±100 scale. If you’ve ever wished RSI reacted quicker, the CMO is worth knowing.

The one-sentence job

The CMO measures net momentum by comparing the sum of recent up-moves to the sum of recent down-moves, as a percentage of total movement. All buying and no selling reads +100; all selling reads −100; balanced reads 0.

The formula

Over the last N bars (Chande’s default is 9):

1. For each bar, take the change from the previous close.
2. SU = sum of all the UP changes over N bars
3. SD = sum of the absolute value of all the DOWN changes over N bars
4. CMO = 100 × (SU − SD) / (SU + SD)

The numerator, SU − SD, is net momentum: positive when buyers dominated, negative when sellers did. The denominator, SU + SD, is total movement, which normalizes the result onto the −100 to +100 range regardless of the instrument’s price. Simple, symmetric, and honest.

How it differs from RSI (the whole point)

RSI and CMO both compare up-moves to down-moves, but two design choices set the CMO apart:

  • No Wilder smoothing. RSI runs its average gain and average loss through Wilder’s exponential smoothing. The CMO uses plain sums over the window — no smoothing at all. That makes it react faster and reach the extremes more often.
  • It uses both up and down in the numerator. RSI’s formula effectively looks at the ratio of average gain to average loss; the CMO puts SU − SD directly on top. The effect is a cleaner measure of net directional pressure.

The practical upshot: the CMO is more responsive and more volatile than RSI. It gives earlier signals at the cost of more false ones — the classic speed-versus-noise trade-off.

How to read it

Overbought / oversold. Because it’s faster, the CMO uses wider thresholds than RSI: +50 for overbought and −50 for oversold are the common levels (some traders use ±40 or ±60). Above +50, buyers have been strongly in control; below −50, sellers.

Zero line. Crossing above zero means up-moves now outweigh down-moves over the window — a momentum flip. The zero cross is the CMO’s simplest trend-direction cue.

Divergence. As with any oscillator, price making a new high while the CMO makes a lower high warns that momentum is fading beneath the move.

Trend strength. Chande also intended the CMO as an input to adaptive systems — for example, feeding its absolute value into a variable-length moving average so the average speeds up when momentum is strong. That’s a more advanced use, but it’s why he built it.

Where it lies to you

  • Faster means noisier. The lack of smoothing that makes the CMO responsive also makes it whipsaw in choppy markets. Wider thresholds (±50) exist for a reason.
  • Extremes aren’t reversals. A CMO pinned near +100 signals a strong trend, not an imminent top. Fading it blindly is how you get run over.
  • Short default is twitchy. The 9-period default is quick; bump it to 14 or 20 for smoother, less frequent signals.
  • It’s momentum, not price. Like all oscillators, it can diverge from price for a long time before anything happens.

The CMO is a clean, symmetric momentum reading that trades RSI’s smoothness for speed. If you like RSI’s logic but find it slow, this is the tool to test.

Now go test it, don’t trust it

Does a CMO zero-cross, or a ±50 reversal, actually beat the equivalent RSI rule on your instrument? That’s a direct A/B test. Build both in AlgoGen. Then compute the CMO yourself in Python, MQL5, Pine Script, EasyLanguage, or NinjaScript.


This post is educational, not financial advice. Indicators describe the past; they don’t predict the future. Backtest anything before you risk real money on it.

Historical research from the Algogen archive. Not investment advice.

Blog

Have a strategy idea?

Describe it in plain English, preview where your rules fire, then decide whether the evidence is worth a backtest.

Keep me postedHow it works